Basically I think it is ok to blaze through the opening as long as familiar moves are being played or everything is "in book". But when your opponent plays something different then it's time to SLOW DOWN and see if their move is safe or how you might refute it. See how the move changes the position from the tabiya and it could be their move was a mistake.

I like to think that there are two opening phases in a chess game.

The first opening is when I am operating purely on memory. Then I know I am safe. I can bang the moves out with confidence as if I were Kasparov or Casablanca reborn. My 1. e4 is the stuff of legend. As Irving Chernev said in Logical Chess, you can play this move with certainty that no grandmaster alive or dead could play a better move.

The second opening starts when my opponent plays something that I do not know. Then I have to think for myself. Now comes the time for thinking and watching out for traps. If my opponent has played something out of book, then maybe he has made a mistake that could be exploited? Or maybe he knows the opening better than I do?

But other than that, this second opening is a time to develop pieces, don't go grabbing pawns, castle, and decide where the rooks are going to go.

This phase ends when the opening becomes a middlegame, ie when the pawn structure is more or less established and the rooks have something to do.

GMs have long long first openings, sometimes all the way into a middlegame or even an endgame. Us patzers quickly find ourselves in phase 2.

That's why I like opening books with lots of text to explain the middlegame plans. Okay, I've run out of memorised moves. Now what do I do?

The scene: a wood somewhere in Germania. The Roman army are arranged for war - catapults to the rear, then archers, then the foot soldiers. Nearby, the cavalry are hidden by thicket of trees. Maximus picks up a handful of earth and rubs it between his fingers to remind himself of his farm in Spain where his wife and son wait for him.

Suddenly, a horse gallops from the far side of the wood, its headless rider tied to the saddle.

"They say no," says Maximus, with a hint of sadness in his voice. He has seen too many battles to be either afraid or to relish another.

"Some people don't know when they're beaten," says his lieutenant.

The pawn at e5 is immune owing to 5...♕a5+ picking off the bishop.

I figured that ...Qa5+ was probably part of the solution, although playing the immediate 4...Qa5+ doesn't work so well, since 5.Bd2 or 5.Qd2 or 5.Nc3 or 5.c3 all get white out of trouble (note that 5...e5 can be answered with 6.N4b3 attacking the queen.)

However, the queen check would have teeth if it could attack another piece. As soon as that idea came to mind, I spotted the pawn fork <4...e5> as a decoy that sets up the queen fork: 5.Bxe5 and *now* Qa5+ scoops up the bishop for a pawn.

e.g. after 4 ... e5, the conversation might have gone ....

Oscar: Oh dear, I've wasted a piece. Indeed, my situation is serious, but it's not hopeless. I'll play on.

Black to play (4...?) "Easy"

4.Nxd4 -1.90 was the first of White's blunders (Qxd4 or Bxb8 better) culminating with 19.Qf1 -13.10. Qc1, -3.65, would have been the best
futile survival effort.